LinkBack URL
About LinkBacks
My advice is DON'T DO IT!!!!Originally Posted by sdantonio
I'd hang on until MrBean invents a frikin laser beam (laser bean?) scanner device
...then again
| | |
Shallow would be OK. 15-20 degree maybe?
I didn't use weight. The spring is pretty strong now so don't think it will have much effect.
I think it is pushing a bit lop-sided though. Thats why i think 3 separate springs would be better.
My advice is DON'T DO IT!!!!
I'd hang on until MrBean invents a frikin laser beam (laser bean?) scanner device
...then again
Have a question:
When using the probe is it necessary to move the probe by hand, or could you define the area and then have it probe the piece on a .X mm grid automatically?
Thanks
Mach3 has a couple options, one of which is a wizard that let's you specify the grid size and spacing.
Gerry
Mach3 2010 Screenset
http://home.comcast.net/~cncwoodworker/2010.html
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
That's right. It's pretty basic stuff but it does the job.
It actually generates a probing program in g-code which does the scanning.
Biggest problem with it is speed (I think there may also be issues with larger numbers of data points).
Having to probe down, retract, move along a fraction of a mm then probe down again a few thousand times is slow going. Because the distances are quite small you rarely get out of the 'ramping' area of the motor speed control.
Because these probes work sideways as well, a more intelligent method would scan down then zig-zag until it hits something.
A multi-pass thing would be good to. First pass to get rough dimensions then a finer pass to digitize details. This way you can spend more time at faster feed rates.
I think it would be a nightmare to do something like this in g-code. I'd like to knock up a stand alone program to do this.
@*!?!& and $??#!@-ing %*!!?-ards!!!!
Things were going so well, then open circuits started to happen again.
I'm sure with a few tweaks they'll go away.
This is a 0.2mm grid scan of a dog tag. This took ~7 hours to scan.
I think I'll put probing on hold for a bit until I get more time to play with the mechanics and write some scanning software. (I've got 16 8' x 4' 18mm birch ply sheets coming in the next couple of days so I may be a little preoccupied).
I like the scanning idea.
If one could set aside enough memory for a completed scan of an object, taking into consideration the size and final resolution of the image, one might be able to speed up the scan process.
Think of it like a topographical map. It is scanned at Z height. If any peaks appear at that height, they should activate the probe and those points could be located at the correct X, Y, and Z points in memory.
On the next pass, the Z height is lowered. The probe is looking for contact. Maybe a look Ahead function could also be worked out where the system knows where the previous contact points were, and can bypass those but continue the scan at the next point.
This would allow the probe to move at some speed at a fixed point in Z axis space until contact is made. If the look ahead function works, the previously scanned areas would be bypassed and only the areas without data would be continiously probed until the scanning was complete.
Guys, I know this is complex, and I left out a lot of steps, but it might be possible if someone has a few hundred hours to spend on it.
Jerry
All the commercial scanners and computer controlled measuring systems uses tools with ball tips. Shouldn't be a big problem to put a ball bearing ball at the end of the tip.My advice is DON'T DO IT!!!!
I'd hang on until MrBean invents a frikin laser beam (laser bean?) scanner device
...then again
although im sure physical measuring is used for a lot of tasks, it really isnt suitable for use on antiques, as 10bulls mentioned. I have previously used laser scanners to image small statues etc, but these tend to be very expensive.
As for a 'cheap' diy solution, well if you consider a dlp projector and a decent digicam cheap, you can do pretty well with structured light scanning. Its good to sub-mm accuracy, but not as accurate as a touch probe. http://www.ercim.org/publication/Erc.../rocchini.html
Playing around with this has made me want to try a more sensitive probe. I was thinking of using probe tips epoxied to piezzo transducers. A gentle poke should give a good voltage spike to pick up.
...or even use a record player stylus!
I couldn't resist having one last tweak tonight. I manually bent the brass tube pins upwards and put 3 separate springs in. I put a bit of double sided carpet tape over each pin. This holds them down, stops then sliding AND gives something for the spring to stick to so it doesn't sproing out.
I also put quite long m3 screws that hold the cap down. This allows for a wide range of adjustment. Tight down results in quite a hard poking force but no doubt reliable. A couple of turns loosened I managed to get it to a point where I could reliably trigger it with my finger without drawing blood and it
didn't seem to stick. Though I don't think I'd trust it to scan any part of my anatomy any time soon.
I'm going to give it one last go tomorrow for now.
Last edited by 10bulls; 12-12-2006 at 07:17 PM.
Another use for a touch probe.
Having a stiff probe may not be so bad. If you had a pointy diamond tip on the probe, you could record how far you had to push to trip the probe.
You could equate this to hardness (maybe).
At metallurgy work experience we did something similar but measured the width of the diamond imprint with a microscope).
I've often wondered how well my experiments at tempering metal work. You could use this as some sort of indicator.
It could also make a pretty effective body piercer in the wrong hands.
I have been measuring prototypes in clay with mechanical devices. The CE-J machines are of course extremely sensitive, but if you need to leave marks on the measured device, the probe spring is too strong.
That's at least my opinion.
Regards,
Sven
Posting Permissions
| | |
